]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Greg Farnum [Wed, 7 Dec 2011 18:56:01 +0000 (10:56 -0800)]
SimpleMessenger: remove void send_keepalive.
Nobody uses this; they all call the version that returns an int.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Mon, 12 Dec 2011 17:29:36 +0000 (09:29 -0800)]
Merge branch 'wip-mon-stats'
Sage Weil [Thu, 8 Dec 2011 21:13:29 +0000 (13:13 -0800)]
mon: add mds, mon info to cluster_logger
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 8 Dec 2011 21:02:12 +0000 (13:02 -0800)]
mdsmap: rename get_num_*_mds() methods
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 8 Dec 2011 20:53:14 +0000 (12:53 -0800)]
mon: report basic cluster stats via perfcounters
These are basic point-in-time cluster stats.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 12 Dec 2011 17:22:49 +0000 (09:22 -0800)]
crush: drop unused label
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 12 Dec 2011 17:20:26 +0000 (09:20 -0800)]
Merge remote branch 'gh/stable'
Sage Weil [Wed, 7 Dec 2011 17:07:38 +0000 (09:07 -0800)]
crush: fix force to behave with non-root TAKE
If the (first) TAKE in the crush rule is not the root, see if they picked
a point somewhere beneath the appropriate part of the hierachy. If not,
continue with a normal mapping.
Previously we would simply error out in this case and map nothing.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 7 Dec 2011 17:09:32 +0000 (09:09 -0800)]
crush: simplify force argument check
force isn't used past this point, only force_pos. Collapse the if
conditions.
Signed-off-by: Sage Weil <sage@newdream.net>
Josh Durgin [Sat, 10 Dec 2011 00:31:07 +0000 (16:31 -0800)]
workunits: print tests pjd runs
This will tell us which ones actually failed within a test suite.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Samuel Just [Fri, 9 Dec 2011 23:23:32 +0000 (15:23 -0800)]
Merge branch 'wip_pgls'
Samuel Just [Fri, 9 Dec 2011 22:42:21 +0000 (14:42 -0800)]
pybind: add object locator support to pybind pool listing
list_objects returns Object(). Object therefore now has an optional
locator_key parameter which will set up the object locator on Object()
methods to allow objects returned from list_objects with locator keys to
be used normally.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Thu, 8 Dec 2011 01:06:01 +0000 (17:06 -0800)]
rados.cc: add --object-locator and object locator output to ls
--object-locator locator causes io to use the specified locator. For
objects with non-empty locators, rados pool ls will output the locator
as well.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Sat, 3 Dec 2011 01:06:59 +0000 (17:06 -0800)]
ReplicatedPG: collection_list_handle_t is now an hobject_t
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Fri, 2 Dec 2011 23:42:01 +0000 (15:42 -0800)]
osd: delay pg list on a snapid until missing is empty
We cannot determine from the missing set whether an object existed
at a given snap.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Greg Farnum [Fri, 9 Dec 2011 01:45:47 +0000 (17:45 -0800)]
msgr: turn up socket debug printouts
These shouldn't be too common and will help in debugging
socket leaks.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Josh Durgin [Fri, 9 Dec 2011 00:36:45 +0000 (16:36 -0800)]
udev: drop device number from name
The device number depends on how many rbd images have been
mapped. Removing it makes the name determined solely by the name,
image, and snapshot that are mapped, for ease of scripting or persistence
across reboots.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Henry C Chang [Thu, 8 Dec 2011 14:35:07 +0000 (22:35 +0800)]
pybind: trivial fix of missing argument
Signed-off-by: Henry C Chang <henry.cy.chang@gmail.com>
Sage Weil [Thu, 8 Dec 2011 16:42:59 +0000 (08:42 -0800)]
crush: whitespace
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 16:41:54 +0000 (08:41 -0800)]
osdmap: initialize cluster_snapshot_epoch
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 16:41:39 +0000 (08:41 -0800)]
crush: set max_devices=0 for map with empty buckets
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 16:01:12 +0000 (08:01 -0800)]
crush: fix stepping on unallocated memory
If size is 0 we can't write here.
Reported-by: pankaj singh <psingh.ait@gmail.com>
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 05:16:18 +0000 (21:16 -0800)]
client: trim lru after flushing dirty data
Shouldn't matter, but it would be interesting to see if this affects
#1737.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 05:05:40 +0000 (21:05 -0800)]
client: unmount cleanup
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 05:01:48 +0000 (21:01 -0800)]
client: wait for sync writes even with cache enabled
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 8 Dec 2011 04:59:25 +0000 (20:59 -0800)]
client: send umount warnings to log, not stderr
stderr isn't usually open anyway.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Samuel Just [Fri, 2 Dec 2011 21:27:43 +0000 (13:27 -0800)]
ObjectStore,ReplicatedPG: remove old collection_list_partial
No need for the old collection_list_partial instance: it's cleaner to
just use an hobject_t as the collection list handle.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Thu, 1 Sep 2011 21:56:01 +0000 (14:56 -0700)]
librados,Objecter,PG: list objects now includes the locator key
Previously, there was no way to recover the locator key used to create
and object. Now, rados_objects_list_next and ObjectIterator will return
the key as well as the object name.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Mon, 5 Dec 2011 19:25:09 +0000 (11:25 -0800)]
hobject_t: encode max properly
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 1 Dec 2011 21:12:43 +0000 (13:12 -0800)]
hobject_t: make filestore_hobject_key_t 64 bits
So we can return 0x100000000 when max=true.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 1 Dec 2011 19:30:50 +0000 (11:30 -0800)]
os/HashIndex: some minimal debug output
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 04:57:23 +0000 (20:57 -0800)]
ObjectStore: Add collection_list_partial for hash order
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Thu, 1 Dec 2011 18:58:37 +0000 (10:58 -0800)]
hobject_t: define explicit hash, operator<<; drop implicit sobject_t()
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Samuel Just [Wed, 30 Nov 2011 18:38:09 +0000 (10:38 -0800)]
object.h: Sort hobject_t by nibble reversed hash
To match the HashIndex ordering, we need to sort hobject_t by the nibble
reversed hash. We store objects in the filestore in a directory tree
with the least significant nibble at the top and the most at the bottom
to facilitate pg splitting in the future.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 04:57:09 +0000 (20:57 -0800)]
hobject_t: sort by (max, hash, oid, snap)
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 04:56:50 +0000 (20:56 -0800)]
hobject_t: define max value
Create a max value that is greater than all other values.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Tommi Virtanen [Tue, 6 Dec 2011 23:22:16 +0000 (15:22 -0800)]
gitignore: Ignore src/keyring, as created by vstart.sh
Commit
86c34ba9ee8c883b71a8449c3c261154365c35ae changed
the filename but not .gitignore.
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Samuel Just [Tue, 6 Dec 2011 21:23:03 +0000 (13:23 -0800)]
ReplicatedPG: don't crash on empty data_subset in sub_op_push
If data_subset is empty (i.e., the data we pulled is no longer useful),
we should mark complete false and continue rather than fail the
assert in range_end().
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Josh Durgin [Tue, 6 Dec 2011 22:19:19 +0000 (14:19 -0800)]
workunits: fix secret file and temp file removal for kernel rbd
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Greg Farnum [Tue, 6 Dec 2011 22:24:08 +0000 (14:24 -0800)]
ReplicatedPG: do not ->put() scrub messages when adding to a WorkQueue.
This function is passing a reference from PG::active_rep_scrub to
the req_scrub_wq, not eliminating the reference (and the WorkQueue
doesn't grab a new reference itself, either).
The other alternative is to convert the WorkQueue to grab a
reference, but since they can cycle through the WorkQueue more than
once, and need to be ->put() outside the WorkQueue, I don't like
that option.
This should fix #1758.
Also add an assert to PG::_request_scrub_map to check on the other
possible cause of this bug (and fix the indentation).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Josh Durgin [Tue, 6 Dec 2011 21:36:51 +0000 (13:36 -0800)]
workunits: make rbd kernel workunit executable
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Tommi Virtanen [Tue, 6 Dec 2011 20:13:03 +0000 (12:13 -0800)]
doc: Reorganize pip calls to use a requirements file.
The conditional before running pip install was unnecessary,
"pip install" on already installed packages is fast (as long
as it's not --upgrade), and --quiet makes it not spam the
console.
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Tommi Virtanen [Tue, 6 Dec 2011 20:04:03 +0000 (12:04 -0800)]
doc: Switch diagram tools from dia to ditaa.
Now you can create diagrams easily with the ".. ditaa::"
directive in the Sphinx documents.
admin/build-doc now checks for debs required for building
the documentation, or just lists commands missing for hosts
not using dpkg.
For more on Ditaa, see http://ditaa.sourceforge.net/
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Sage Weil [Tue, 6 Dec 2011 18:46:27 +0000 (10:46 -0800)]
filestore: send back op error to log, not stderr
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 6 Dec 2011 18:39:49 +0000 (10:39 -0800)]
doc: fix typo
Signed-off-by: Sage Weil <sage@newdream.net>
Josh Durgin [Tue, 6 Dec 2011 18:31:17 +0000 (10:31 -0800)]
workunits: add some tests for kernel rbd
This covers some snapshot and resize functions that aren't tested by fs benchmarks.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Wed, 30 Nov 2011 18:26:22 +0000 (10:26 -0800)]
rbd: the showmapped command shouldn't connect to the cluster
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Tue, 29 Nov 2011 01:24:25 +0000 (17:24 -0800)]
ceph-rbdnamer: include snapshot name if present
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Tue, 29 Nov 2011 01:02:15 +0000 (17:02 -0800)]
rbd, mount.ceph: use pre-stored secret if available
If a secret is specified, store and use it, but otherwise
check for a pre-existing secret to use.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Tue, 29 Nov 2011 00:54:27 +0000 (16:54 -0800)]
secret: add is_kernel_secret function
This will let us know whether we can add a key mount option
if no secret is specified.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Tue, 29 Nov 2011 00:52:19 +0000 (16:52 -0800)]
secret: fix error check
add_key will return -1 when an error occurs, which should be handled at a higher level and not printed here.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Wed, 23 Nov 2011 21:54:08 +0000 (13:54 -0800)]
rbd: allow snapshots to be mapped
unmap and showmapped already support snapshots. map should too.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Tue, 6 Dec 2011 16:34:19 +0000 (08:34 -0800)]
test_rados.py: clean up after EEXIST test
This extra pool caused subsequent pool tests to fail.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Sage Weil [Tue, 6 Dec 2011 01:33:57 +0000 (17:33 -0800)]
Merge remote branch 'gh/stable'
Sage Weil [Tue, 6 Dec 2011 00:16:35 +0000 (16:16 -0800)]
doc: fix rst syntax
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 5 Dec 2011 22:07:44 +0000 (14:07 -0800)]
doc: document monitor cluster expansion/contraction
Pretty sure my rst syntax is wrong.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 5 Dec 2011 21:38:02 +0000 (13:38 -0800)]
cephtool: fix shutdown
Fix 'ceph -w' brokenness from commit
ad13d0b7 .
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Mon, 5 Dec 2011 17:36:54 +0000 (09:36 -0800)]
filejournal: make FileJournal::open() arg slightly less weird
Pass in fs_op_seq (last_committed_seq), not the next expected seq, so we
can avoid subtracting and adding 1 in odd places.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 5 Dec 2011 19:21:08 +0000 (11:21 -0800)]
Merge branch 'stable'
Sage Weil [Mon, 5 Dec 2011 17:23:56 +0000 (09:23 -0800)]
vstart.sh: .ceph_keyring -> keyring
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 5 Dec 2011 18:52:24 +0000 (10:52 -0800)]
filejournal: remove bogus check in read_entry
It is perfectly fine to read events that are older than the fs's seq from
the journal; open() will skip them when positioning the read pointer on
open.
Also, this code is nonsensical; it always failed the assertion.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 5 Dec 2011 17:34:44 +0000 (09:34 -0800)]
filejournal: set last_committed_seq based on fs, not journal
last_committed_seq is the last seq committed to the fs, not the journal.
Set it when we begin replay with the fs provided value, not from the newest
entry in the journal.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 2 Dec 2011 23:35:38 +0000 (15:35 -0800)]
mon: stub perfcounters for monitor, cluster
The 'mon' perfcounter is for the local daemon and is always registered.
The 'cluster' perfcounter is for cluster state, and is only registered
(and thus only shows up via the admin socket) when the current daemon is
part of the cluster quorum.
No actual counters yet.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 2 Dec 2011 23:25:37 +0000 (15:25 -0800)]
osd: safely requeue waiting_for_ondisk waiters on_role_change
This could conceivably cause the reply ordering mismatch seen in bug
#1490. Not sure why we didn't also fix this caller when we fixed that
bug last time :).
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 2 Dec 2011 23:24:00 +0000 (15:24 -0800)]
osd: rename {take -> requeue}_object_waiters
It calls osd->requeue_ops(), so make naming more consistent and avoid
confusing people like me.
Signed-off-by: Sage Weil <sage@newdream.net>
Josh Durgin [Fri, 2 Dec 2011 21:17:34 +0000 (13:17 -0800)]
rados.py: add list_pools method
Signed-off-by: Eric Chen <Eric_YH_Chen@wistron.com>
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 20:06:50 +0000 (12:06 -0800)]
Merge branch 'stable'
Mark Kampe [Fri, 2 Dec 2011 19:26:20 +0000 (11:26 -0800)]
Doc: add a conceptual overview of the peering process
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 19:19:21 +0000 (11:19 -0800)]
mds: remove obsolete doc
Sage Weil [Fri, 2 Dec 2011 17:58:45 +0000 (09:58 -0800)]
crush: ignore forcefed input that doesn't exist
This might happen if, e.g., the file_layout specifies an osd that later
is removed from the cluster entirely. Just ignore it instead of making
upper layers duplicate this check.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 17:47:58 +0000 (09:47 -0800)]
Revert "CrushWrapper: ignore forcefeed if it does not exist"
This reverts commit
6fbab6da6942c238d40a6b4f1680a7e6da463289 .
This fails a unit test.
And I change my mind.. I think this is most cleanly handled inside crush, so
we don't duplicate the same check that is generating the error with an different
data structure.
Sage Weil [Fri, 2 Dec 2011 17:01:31 +0000 (09:01 -0800)]
v0.39
Samuel Just [Fri, 2 Dec 2011 00:28:03 +0000 (16:28 -0800)]
OSDMap: build_simple_from_conf pg_num should not be 0 with one osd
Previously, pg_num would end up set to 0 if osd.0 is the only osd.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Fri, 2 Dec 2011 04:36:13 +0000 (20:36 -0800)]
objecter: initialize global_op_flags to zero
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Mark Kampe [Thu, 1 Dec 2011 23:58:32 +0000 (15:58 -0800)]
Doc: delete gratuitous index.html
It was not an index, and seems to contain recommendations
for system configuration. I have renamed it to confusing.txt
and will merge it in a future commit.
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Mark Kampe [Thu, 1 Dec 2011 23:34:57 +0000 (15:34 -0800)]
Doc: complete reversion of architecture.rst
(abandon in progress improvements until everything works)
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Mark Kampe [Thu, 1 Dec 2011 23:29:00 +0000 (15:29 -0800)]
Doc: deleted gratuitious PlanningImplementation.html,
which was a copy of PlanningImplementation.txt
(and not html at all).
restored previous index.rst, which was overwritten with a copy
of PlanninImplementation.txt, but removed all of the recursively
included content from the document.
I will cherry-pick merge the new contents in a subsequent commit.
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Mark Kampe [Thu, 1 Dec 2011 23:22:15 +0000 (15:22 -0800)]
Doc: Restore the previous version of architecture.rst
it was accidentally overwritten with a version of the product
had a somewhat different audience/focus and a few sphinx
formatting errors.
I will cherry-pick the corrections in a subsequent commit.
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Mark Kampe [Thu, 1 Dec 2011 22:59:24 +0000 (14:59 -0800)]
doc: change state model from .svg to .png
Signed-off-by: Mark Kampe <mark.kampe@dreamhost.com>
Steve MacGregor [Mon, 31 Oct 2011 19:31:41 +0000 (15:31 -0400)]
fixed ubuntu version typo
Samuel Just [Tue, 29 Nov 2011 01:30:37 +0000 (17:30 -0800)]
CrushWrapper: ignore forcefeed if it does not exist
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Josh Durgin [Tue, 15 Nov 2011 22:27:53 +0000 (14:27 -0800)]
librbd: report an error if rbd header does not match
This will fail on future incompatible versions of the header format.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Greg Farnum [Thu, 1 Dec 2011 19:15:44 +0000 (11:15 -0800)]
Merge branch 'wip_local_reads'
Greg Farnum [Thu, 1 Dec 2011 19:14:34 +0000 (11:14 -0800)]
hadoop: apache license.
We haven't made explicit that the Hadoop Java code is under the Apache
License. Do so (with permission from the other contributors, thanks!).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Thu, 1 Dec 2011 17:17:38 +0000 (09:17 -0800)]
mds: fix blocking in standby replay thread
We need to hold mylock before waiting on the cond or else we get
./common/Cond.h: In function 'int Cond::Wait(Mutex&)', in thread '
7f37fe0c8700 '
./common/Cond.h: 46: FAILED assert(mutex.is_locked())
ceph version
0.38-2-g73f99a1 (commit:
73f99a189f491866da2be88adcfe0bd512282755 )
1: (MDLog::_replay_thread()+0x2483) [0x6c4393]
2: (MDLog::ReplayThread::entry()+0xd) [0x4decbd]
3: (()+0x6d8c) [0x7f3803e8fd8c]
4: (clone()+0x6d) [0x7f38028d504d]
ceph version
0.38-2-g73f99a1 (commit:
73f99a189f491866da2be88adcfe0bd512282755 )
1: (MDLog::_replay_thread()+0x2483) [0x6c4393]
2: (MDLog::ReplayThread::entry()+0xd) [0x4decbd]
3: (()+0x6d8c) [0x7f3803e8fd8c]
4: (clone()+0x6d) [0x7f38028d504d]
*** Caught signal (Aborted) **
in thread
7f37fe0c8700
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 1 Dec 2011 17:17:00 +0000 (09:17 -0800)]
global: make daemon banner print explicit
This eliminates some flags and avoids annoying cases where the banner is
printed but we don't want to see it.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 1 Dec 2011 16:19:47 +0000 (08:19 -0800)]
mds: fix usage text
Filename is not optional.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 30 Nov 2011 17:57:29 +0000 (09:57 -0800)]
mds: adjust flock lock state on export
Looks like this was missed when flocklock was added. Did a quick grep and
it doesn't look like it is missing anywhere else.
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Wed, 30 Nov 2011 02:14:29 +0000 (18:14 -0800)]
Objecter: loop the right direction when searching for local replicas
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Samuel Just [Wed, 30 Nov 2011 00:24:35 +0000 (16:24 -0800)]
doc: Add peering state diagram
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Tue, 29 Nov 2011 23:36:07 +0000 (15:36 -0800)]
Makefile: ipaddr.h, pick_address.h
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 29 Nov 2011 21:31:38 +0000 (13:31 -0800)]
Makefile: add missing uuid.h to tarball
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Tue, 29 Nov 2011 21:30:45 +0000 (13:30 -0800)]
Objecter: fix local reads in recalc_op_target
We want to use the actual OSD, not the index into the array!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Tue, 29 Nov 2011 16:28:57 +0000 (08:28 -0800)]
osd: subscribe to next map if flagged FULL
This ensures the osd finds out when we become un-full in a timely manner.
Fixes: #1755
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Tue, 29 Nov 2011 05:37:18 +0000 (21:37 -0800)]
mds: encode truncate_pending in inode
Otherwise we don't actually journal this value, and we get confused when
we replay a start_truncate and try to restart it.
Fixes: #1756
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Greg Farnum [Mon, 28 Nov 2011 20:45:09 +0000 (12:45 -0800)]
uclient: remove filer_flags and use Objecter::global_op_flags instead
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 28 Nov 2011 20:42:21 +0000 (12:42 -0800)]
Objecter: add a new global_op_flags that is passed to every Op constructor.
We can use this for a global use of LOCALIZE_READS (and are about
to do so!).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 28 Nov 2011 20:30:46 +0000 (12:30 -0800)]
Objecter: remove unused variable in op_submit
These flags are probably relics from when the function got split;
they belong in send_op now.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 28 Nov 2011 18:32:07 +0000 (10:32 -0800)]
uclient: remove useless if-else based on snapid
These are the same command anyway!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Wido den Hollander [Wed, 16 Nov 2011 19:41:15 +0000 (20:41 +0100)]
debian init: Do not stop or start daemons when installing or upgrading
Signed-off-by: Wido den Hollander <wido@widodh.nl>